ApplicationNo. 06/302590 filed on 09/15/1981
US Classes:623/3.21, Flexible chamber or diaphragm dirrectly compressed by pressurized working fluid417/383, Pulsator or fluid link417/394, Collapsible common member417/395Diaphragm
ExaminersPrimary: Frinks, Ronald L.
Attorney, Agent or Firm
International ClassesA61M 1/10 (20060101)
A61M 1/12 (20060101)
DescriptionBACKGROUND OF THE INVENTION
This invention relates to the field of cardiac prosthetic devices, and, more particularly, to hydraulically actuated total replacement artificial hearts and circulatory assist devices, including left ventricular assist devices, especially for useby and implantation in humans.
It has been estimated that between 16,000 and 50,000 patients annually are suitable candidates for implantation of a total cardiac prosthesis (TCP). Such candidates typically are disabled due to insufficient left and right ventricular functionbut are otherwise in good health. Many thousands more annually with inadequate left ventricular function and satisfactory right ventricular function may be candidates for a permanently implanted left ventricular assist device (LVAD).
The ideal total cardiac prosthesis must provide complete rehabilitation for the patient. Such a TCP recipient must be able to engage in gainful employment and all normal activities including moderate exercise. He should retain a substantiallynormal appearance and normal or near normal mobility with no significant limitations of any kind. Cardiac output effected by the TCP must be normal, adequate and sufficiently responsive to the patient's requirements to accommodate expected, suddenchanges in physical activity or emotional stress level. The presence and operation of the TCP must be sufficiently unobtrusive so that the patient can largely forget that he is dependent on an artificial heart. All blood pumping functions of the TCPshould be completely automatic, so that the patient performs no control or monitoring functions except for maintaining adequate power to the TCP, and responding to warnings that indicate a lack of power or serious problems requiring immediate technicalor medical attention.
The intrathoracic blood pumping components of the TCP must be similar in size and weight to the natural heart. TCP life must be sufficiently long and reliability sufficiently high that risk to the patient of sudden prosthesis failure and itsattendant anxiety are minimized. The formation of pannus and adherent thrombus must be prevented to avoid a compromise of blood pump function. Thrombo-emboli and excessive blood damage also must be prevented. The TCP must not damage adjacent tissuesor impair organ function by toxicity and adverse tissue reactions, by mechanical trauma or compression, or by excessive local temperatures. The system must avoid skin penetrations of any kind to prevent infections that can arise from percutaneous leads. This eliminates a major risk to the patient, reduces the need for clinical observation and treatment, and reduces the maintenance of the TCP required of the patient. This ideal system must be low in cost to purchase, implant, and maintain. Thefrequency and extent of routine monitoring and maintenance, both medical and technical, must be low.
Serious research toward the realization of a total cardiac prosthesis has been under way since about 1957, sponsored largely by the U.S. National Institutes of Health (NIH). Researchers have directed this activity to six principal areas: (1)blood-compatible materials for the blood pumping means; (2) heart valves; (3) blood pumps; (4) blood pump actuating means; (5) power supplies and their application to the internal blood pump actuating means; and (6) control mechanisms for the pumpingfunction.
Many materials have been developed which apparently achieve blood compatibility. See, e.g., the recent survey and evaluation of these by M. Szycher et al in "Selection of Materials for Ventricular Assist Pump Development and Fabrication", Trans. ASAIO, Vol. XXIII, p. 116, 1977, incorporated herein by reference. (As used herein, Trans. ASAIO refers to the Transactions American Society For Artificial Internal Organs). While there is yet no human experience, recent materials like Biomer,Avcothane, etc., have been benign (i.e., have not caused thrombo-emboli) for periods up to 221 days in the calf. The materials for blood pumping membranes or sacs, however, must not only be benign and tissue compatible, but also able to withstand tensof pounds of force for something on the order of 109 flexing cycles during a 20-year prosthesis life. Apparently, appropriate materials are nearly, if not already, realized today.
Another critical element of the blood pumping means is the valves, which permit blood flow into or out of the heart, but prevent backflow of blood. Many different types of valve prostheses have been developed and used in tens of thousands ofimplants to replace defective natural valves. Hence, adequate valves for a TCP appear to be well within the state of the art.
There has been a great deal of development activity in the area of blood pumps, primarily associated with LVAD's. This experience has shown that by utilizing appropriate biocompatible materials as described above, adequate and reliable bloodpumps can be designed. The most common form of blood pump is comprised of an elastomeric sac or diaphragm-capped cavity. In a TCP which comprises two such blood pumps, each cavity is fitted with an inlet valve and an outlet valve. These pumpingcavities replicate the function of the adjacent right and left ventricles of the natural heart.
The least developed of the aforementioned areas of activity is the development of an actuator to couple the power supply to the blood pump. In order to squeeze the blood-pumping sacs or force the diaphragm into the blood pumping cavity,pneumatic actuation means supplied from outside the body are most common. A number of mechanical actuation systems may be found in the literature. All sorts of linkages, gears, cams, etc., have been proposed, but none is known to be successful. Mostof these systems are driven by an electric motor, although some have relied upon piezoelectric devices and other esoteric means. Both copulsation, the technique used by the natural heart, and alternate pulsation of left and right ventricles have beenemployed successfully. See Smith, L. M.; Olson, D. B., Sandquist, G., Grandall, E., Gentry, S., and Kolff, W. J., "A Totally Implantable Mechanical Heart", Proceedings from the European Society of Artificial Organs, Vol. 2, p. 150, 1975. Medicalopinion appears to be impartial regarding this choice.
The coupling of a mechanical drive to the sensitive blood pumping diaphragm or sac is difficult to accomplish without raising excessive stresses and causing fatigue failures. Hence, the preferred coupling means is fluid, either liquid or gas. For example, one group has constructed an electric motor powered, cam-actuated, diaphragm air pump which couples to the blood-pumping sac via pneumatic pressure. See V. Poirier et al, "Advances in Electrical Assist Devices," Trans. ASAIO, Vol. XXIII,p. 72, 1977, incorporated herein by reference. The entirety of the above-described mechanism is intended to be implanted within the thoracic cavity. In the Poirier et al design the motor rotates only once per heartbeat. Because relatively large torqueis required from the motor, it must use strong magnetic fields, employ high current, and is rather heavy.
Burns et al, by contrast, constructed a TCP actuation system using a 10k-40k rpm motor driving a hydraulic pump pressurizing a liquid to actuate the blood-pumping bladders. See W. H. Burns et al, "The Totally Implantable Mechanical Heart, anAppraisal of Feasibility," Annals of Surgery, September 1966, pp. 445-456, and W. H. Burns et al, "The Total Mechanical Cardiac Substitute," Process in Cardiovascular Diseases, Vol. XII, No. 3, 1969, pp. 302-311, both incorporated herein by reference. However, the electromechanically actuated hydraulic switching valve used in this and similar systems to shunt hydraulic fluid back and forth between ventricular actuating chambers has a number of disadvantages. The switching valve itself is relativelylarge and heavy, consumes a great deal of power and is potentially unreliable. Long and large ducts required in this type of system cause undesirable large frictional and inertial losses, and long fluid acceleration times.
Another approach to hydraulic actuation taken by researchers has involved the use of a reversible pump which directly pumps fluid back and forth between the two actuating chambers. See Jarvik U.S. Pat. No. 4,173,796.
On the subject of power, up to this time most TCPs implanted in the calf have been powered pneumatically via transcutaneous tubing into the thoracic cavity. A large external console supplies the proper regimen of pressure variations in order toactivate the internal blood pump. With such a system, calves have lived up to 221 days. Jarvik, "The Total Artificial Heart", Scientific American, Vol. 244, No. 1, pp. 74-80, January, 1981. On another tack, NIH has sponsored considerable effort onthe development of internal nuclear power supplies and, to a lesser extent, of chemical fuel cells. None of this work, however, appears to be promising; in fact, the nuclear effort was terminated by the U.S. Energy Research and DevelopmentAdministration. Additionally, various means of transmitting mechanical power transcutaneously have been attempted, but none appears to be promising. At present, transcutaneous transmission of electricity appears to be the preferred method for poweringa TCP. A second, less preferable, possibility is the supplying of electrical power through percutaneous wire penetrations, but these always pose a threat of infection and are psychologically annoying to the patient.
Several investigators have developed the technique of transcutaneous electrical power transmission. Their approach is to implant a coil under the skin. This coil functions as a transformer secondary winding, receiving power from an inductivelycoupled, external, mating coil juxtaposed therewith to serve as the transformer primary winding. At frequencies on the order of 17 kHz, up to 100 watts have been thus transmitted for many months across the skin of a dog, by Schuder. See, J. C. Schuderet al, "Ultra High Power Electromagnetic Energy Transport Into the Body," Trans. ASAIO, 1971, incorporated herein by reference. Thus, it appears that the inductive delivery across the intact skin of the approximately 30 watts needed to power a TCP iswell within the state of the art.
On the subject of control of a TCP to make it sympathetic to the body, there have been many different approaches and much controversy. Some researchers have attempted to provide no active control. Others have required a control in order toachieve regular beating. See, e.g., W. H. Burns et al, "The Total Mechanical Cardiac Substitute," identified above. Some systems have attempted to control systole (i.e., the contraction phase of the cardiac cycle whose rate is one determinant ofcardiac output) from the left ventricle of the TCP in order to control the systolic pressure in the aorta. Still other systems have attempted feedback control of stroke volume and beat rate.
The natural heart and at least some, if not all, TCPs are comprised of two pumps in series. The right pump receives blood from the vena cava and impels it into the pulmonary artery. The left pump receives blood from the pulmonary vein andimpels blood into the main circulatory system via the aorta. These two pumps must, over time periods considerably longer than that of a few beats, pump nearly the same amount of blood. Otherwise, the delicate pulmonary circuit will either collapse orrupture from a deficiency or excess of blood pumped by the right ventricle relative to the left. Various investigators have included controls in their TCP systems in order to achieve the critical balance between the pumping rate of the right and leftventricles. The major intrinsic mechanism by which the natural heart controls cardiac output is described by Starling's Law, which essentially states that a ventricle will expel during systole essentially that blood which flows into the relaxedventricle during diastole. For the right ventricle, the body controls the "tone", i.e., the pressure in the venous system, so that the pressure in the vena cava (relative to atmospheric pressure) may rise from 5 to 15 mm Hg when there is a demand forhigher blood flow. This pressure change causes approximately a proportional increase in the amount of blood which flows from the vena cava through the tricuspid valve into the relaxed right ventricle during diastole.
It is important to note that the natural heart has no means to suck upon the veins. It can only produce a systolic contraction which expels blood from the ventricular chamber.
Similarly, for the left ventricle, the pressure in the pulmonary vein varies from 5 to 15 mm Hg and produces a proportional increase in blood flow into the left ventricle. If the right ventricle should temporarily pump slightly more than theleft ventricle, the pressure rises in the pulmonary artery, and, as a consequence, in the pulmonary vein, causing more blood to flow into the left ventricle and thereby matching the pumping rate of the left ventricle to that of the right ventricle. Thus, the natural heart achieves the necessary balance between the two pumps in series via simple and direct fluid dynamic means. In a real sense, the heart is the servant, not the master of the circulatory system, and in particular it responds in thefinal analysis to the requirements of the body as reflected by the peripheral oxygen saturation. The above-described intrinsic control can maintain body function even in the absence of extrinsic humoral or neural control.
The body also neurally controls the rate at which the natural heart beats. Cardiac output is a function of the amount of blood ejected during systole, and the rate at which the heart beats. For all but the most strenuous activity, the systolicstroke volume per beat remains substantially constant. Thus, cardiac output is primarily a function of beat rate (i.e., the number of beats per minute). Heart rates can vary from a low of about 40 to as high as 220 beats per minute in a young personand ordinarily from about 60 to 150 bpm in an adult. Cardiac output of the natural heart can vary from about 4 to as high as 24 liters per minute, the latter being the case of a trained athlete. Experience with pacemakers and transplanted naturalhearts shows that beat rate control via neural sensors is unnecessary for a satisfactory life. The hundreds of natural hearts which have been transplanted operate at their own beat frequency, unresponsive to the body's neural demands because there is noneural connection.
The natural control system also ensures that the systolic pressure in the aorta does not drop below about 80 mm Hg, in order to maintain adequate circulation to the brain. The mean pressure in the aorta is established by cardiac output and theperipheral resistance of the vascular systems. In some of the TCPs which previously have been developed, a control means has been provided to maintain pressure in the aorta and atrium within a reasonable range. On the other hand, there is evidence fromnatural heart transplants that such control is unnecessary; transplanted human hearts have no neural connections to the host body and hence their systolic rates are not related to neural control, yet people with such transplants have been able to leadmeaningful lives. It may be concluded that a TCP can be satisfactorily operated without such control. The evidence above teaches that a workable TCP can be made to approximate the natural heart's Starling's Law behavior with relatively simple controloperations.
Thus, a TCP is now technically feasible provided that a competent design is constructed. The critical blood pumping technology appears to be well established and adequate for long-term survival of the recipient. Benign power transmission acrossthe skin can obviate the portent of infection of the thoracic cavity transmitted via percutaneous leads. One major area where satisfactory progress is lacking, however, is the provision of a practical blood pump actuating mechanism. What is needed is asimple, lightweight, reliable, transcutaneously supplied, electrically-driven actuator. This objective is the one to which the present invention is principally adressed.
SUMMARY OF THE INVENTION
Accordingly, it is an object of the present invention to provide a hydraulic actuation system for use in TCPs and circulatory assist devices which obviates many of the drawbacks of the prior art actuation systems.
More specifically, it is an object of this invention to provide a simple and reliable hydraulically actuated cardiac prosthesis, especially a TCP.
Another object of the invention is to provide a total cardiac prosthesis which is substantially unobtrusive and permits the patient to engage in all normal activities, without significant limitations of any kind.
Another object of the invention is to provide a total cardiac prosthesis wherein all blood pumping functions are completely automatic, and wherein minimum patient attention is required to maintain prosthesis operation.
Another object of the invention is to provide a total cardiac prosthesis which permits monitoring of mechanical and physiological information and which is provided with alarms to warn of power failures or other malfunctions.
Another object of the invention is to provide a total cardiac prosthesis which obeys Starling's Law.
These and other objects of the present invention are accomplished in the most basic form of this invention by providing an implantable hydraulic actuation system for supplying motive power to a blood pumping chamber having a flexible portion,comprising an actuation fluid reservoir; actuation fluid pumping means in fluid communication with the reservoir for pumping actuation fluid therefrom; an actuation chamber having an actuation fluid inlet, and an actuation fluid outlet, said actuationchamber being adapted to cause displacement of the flexible portion of said blood pumping chamber in response to changes in volume of actuation fluid in said actuation chamber; and three-way valve means in fluid communication with said actuation chamber,said pumping means and said reservoir, said valve means being adapted to alternately and repetitively direct actuation fluid from said pumping means to said actuation chamber through said actuation fluid inlet, and then from said actuation chamber tosaid reservoir through said actuation fluid outlet, said valve means comprising a fluid-tight inlet manifold on said actuation chamber surrounding said actuation fluid inlet and in fluid communication with said pumping means; a fluid-tight outletmanifold on said actuation chamber surrounding said actuation fluid outlet, said outlet manifold having at least one actuation fluid outlet port in fluid communication with said reservoir; an inlet valve member in said inlet manifold movable between anopen position clear of said actuation fluid inlet to permit actuation fluid entering said inlet manifold from said pumping means to fill said actuation chamber, and a closed position blocking said actuation fluid inlet to prevent the flow of actuationfluid therethrough; an outlet valve member in said outlet manifold movable between an open position clear of said actuation fluid outlet port to permit actuation fluid to drain from said actuation chamber to said reservoir, and a closed position blockingsaid actuation fluid outlet port to prevent flow of actuation fluid therethrough; coupling means interconnecting said inlet and outlet valve members for synchronizing their movements so that when one of said valve members is in its open position theother valve member is in its closed position; and drive means for moving said valve members between their open and closed positions.
The present invention also is directed to an implantable hydraulically actuated blood pumping system comprising a blood pumping chamber having a blood inlet, a blood outlet and a flexible portion; an actuation fluid reservoir; actuation fluidpumping means in fluid communication with the reservoir for pumping actuation fluid therefrom; and actuation chamber having an actuation fluid inlet, and actuation fluid outlet, said actuation chamber being adapted to cause displacement of the flexibleportion of said blood pumping chamber in response to changes in volume of actuation fluid in said actuation chamber; and three-way valve means in fluid communication with said actuation chamber, said pumping means and said reservoir, said valve meansbeing adapted to alternately and repetitively direct actuation fluid from said pumping means to said actuation chamber through said actuation fluid inlet, and then from said actuation chamber to said reservoir through said actuation fluid outlet, saidvalve means comprising a fluid-tight inlet manifold on said actuation chamber surrounding said actuation fluid inlet and in fluid communication with said pumping means; a fluid-tight outlet manifold on said actuation chamber surrounding said actuationfluid outlet, said outlet manifold having at least one actuation fluid outlet port in fluid communication with said reservoir; an inlet valve member in said inlet manifold movable between an open position clear of said actuation fluid inlet to permitactuation fluid entering said inlet manifold from said pumping means to fill said actuation chamber, and a closed position blocking said actuation fluid inlet to prevent the flow of actuation fluid therethrough; an outlet valve member in said outletmanifold movable between an open position clear of said actuation fluid outlet port to permit actuation fluid to drain from said actuation chamber to said reservoir, and a closed position blocking said actuation fluid outlet port to prevent flow ofactuation fluid therethrough; coupling means interconnecting said inlet and outlet valve members for synchronizing their movements so that when one of said valve members is in its open position the other valve member is in its closed position; and drivemeans for moving said valve members between their open and closed positions.
The present invention also provides an implantable hydraulically actuated total cardiac prosthesis comprising a pair of blood pumping chambers each having a blood inlet, a blood outlet and a flexible portion; a pair of actuation chambers eachhaving an actuation fluid inlet, and an actuation fluid outlet, one of said actuation chambers being operatively associated with one of said blood pumping chambers and the other of said actuation chambers being operatively associated with the other ofsaid blood pumping chambers to cause displacement of the flexible portion of said blood pumping chambers in response to changes in volume of actuation fluid in their associated actuation chambers; an actuation fluid reservoir; actuation fluid pumpingmeans in fluid communication with said reservoir for pumping actuation fluid therefrom; and a separate three-way valve means associated with each of said actuation chambers, each of said valve means being in fluid communication with its associatedactuation chamber, said pumping means and said reservoir, and being adapted to alternately and repetitively direct actuation fluid from said pumping means to its associated actuation chamber through said actuation fluid inlet and then from its associatedactuation chamber to said reservoir through said actuation fluid outlet, said valve means comprising a fluid-tight inlet manifold on said actuation chamber surrounding said actuation fluid inlet and in fluid communication with said pumping means; afluid-tight outlet manifold on said actuation chamber surrounding said actuation fluid outlet, said outlet manifold having at least one actuation fluid outlet port in fluid communication with said reservoir; an inlet valve member in said inlet manifoldmovable between an open position clear of said actuation fluid inlet to permit actuation fluid entering said inlet manifold from said pumping means to fill said actuation chamber, and a closed position blocking said actuation fluid inlet to prevent theflow of actuation fluid therethrough; an outlet valve member in said outlet manifold movable between an open position clear of said actuation fluid outlet port to permit actuation fluid to drain from said actuation chamber to said reservoir, and closedposition blocking said actuation fluid outlet port to prevent flow of actuation fluid therethrough; coupling means interconnecting said inlet and outlet valve members for synchronizing their movements so that when one of said valve members is in its openposition the other valve member is in its closed position; and drive means for moving said valve members between their open and closed positions.
BRIEF DESCRIPTION OF THE DRAWINGS
The novel features of the invention are set out with particularity in the appended claims, but the invention will be understood more fully and clearly from the following detailed description of the invention as set forth in the accompanyingdrawings, in which:
FIG. 1 is a block diagram showing the major components of a total cardiac prosthesis system according to the present invention;
FIG. 2 is a sectional view of one type of blood pump suitable for use in a TCP according to the present invention;
FIG. 3 is a schematic illustration of one embodiment of a total cardiac prosthesis actuation scheme according to the invention;
FIG. 4 is a perspective view of one form of the TCP actuation scheme illustrated in FIG. 3;
FIGS. 5-8 and 16 are schematic illustrations of other forms of TCP actuation schemes using three-way valves according to the invention;
FIG. 9 is a front elevational view of a ventricular blood pump incorporating one form of three-way valve according to the invention;
FIG. 10 is a side elevational view of the same;
FIG. 11 is a partial sectional view thereof taken along line 11--11 of FIG. 10;
FIG. 12 is an exploded perspective view of another form of three-way valve according to the invention, showing its relationship to a ventricular blood pump;
FIG. 13 is a side elevational view thereof;
FIG. 14 is a partial sectional view thereof taken along line 14--14 of FIG. 13; and
FIG. 15 is a sectional view thereof taken along line 15--15 of FIG. 13.
The present invention is based at least in part on the discovery of a new approach to the controlled hydraulic actuation of a blood pump for use in cardiac prostheses. This actuation method and the various mechanical forms suitable forpracticing it can be used with equal facility in total cardiac prostheses and circulatory assist devices. Since the similarities of structure and operation of these two classes of devices are well known in the art, the following description of thisinvention will relate primarily to the hydraulically actuated TCP.
The TCP System
FIG. 1 schematically illustrates the basic components of a TCP system according to the invention and the interaction of these components with the physiological systems of the patient. The pulmonary circulation P is maintained by a rightventricular blood pump RV. The systemic circulation S is maintained by a left ventricular blood pump LV. In the preferred embodiment of this invention each of the ventricular blood pumps is powered by a separate hydraulic actuator, RA for the rightblood pump and LA for the left blood pump. The operation of the actuators RA and LA is controlled and monitored by an internal electronic control, power and monitoring circuit C which is powered at times by internal batteries IB. Most of the time,however, power is derived from an external power supply comprising an external battery EB and power circuit PC. External battery EB is rechargeable from a conventional power supply, such as household AC current or automotive DC current. Power isdelivered transcutaneously to the implanted components by magnetic induction from a primary coil C1 to a secondary coil C2.
Preferably, the blood pumps and actuators are implanted within the thoracic cavity, while the internal electronic controls C and internal batteries IB are implanted outside of the thoracic cavity, preferably near the skin so as to permit easyreplacement or servicing of these components by minor surgery. Of course, secondary coil C2 must be located close to the skin for efficient inductive energy transfer.
The blood pumps of the TCP system according to this invention have essentially the same size, configuration and function as the natural heart. These functions include the same stroke volume capability, the same beat rate range, the same atrialfilling pressure range and the same arterial pressure range and profile as in a healthy heart.
Blood pumps suitable for use according to the present invention can be of any of the known designs which are capable of being actuated by hydraulic actuation systems. This class includes systems in which the actuation fluid does work directly ona component of the blood pump as well as those systems in which the hydraulic fluid is coupled to the blood pump by indirect means, such as by magnetic coupling. Of primary interest, however, are those blood pump types in which the hydraulic fluid actsdirectly on a flexible portion of the blood pump. Examples of this type of blood pump include sac-type and membrane-capped cavity types generally known in the art. The preferred blood pumps for use in the TCP of this invention are of the membrane type(sometimes called bladder-type).
These preferred blood pumps essentially comprise a ventricular chamber containing blood inflow and outflow valves. The right and left ventricular blood pumps are generally of the same design except that the housing contain inflow and outflowducts with orientations necessary to achieve appropriate implantability and fit.
One preferred form of blood pump is schematically illustrated in FIG. 2. The blood pump includes a two-piece rigid housing 1 in which is mounted a flexible membrane, or bladder 2, which is fabricated from an elastomeric material. The housingparts 1 and bladder 2 are secured together at their peripheries by a clamping ring 7. As the blood pump fills with blood during diastole, the flexible membrane assumes the position shown in FIG. 2 in solid lines. As hydraulic fluid is added to thehousing on the non-blood side of the membrane (in a manner later described), the change in hydraulic fluid volume causes displacement of the membrane to the position shown in dotted lines. As the membrane is displaced toward the opposite housing wall,blood is forcibly expelled from the blood pump. The membrane should be of such a design that the displacement or deformation occurs uniformly and consistently with each flexing stroke. The membrane also should be designed to intrinsically avoid blooddamaging contact with any portion of the rigid housing, and/or extrinsic control means should be provided to so limit the excursion of the membrane.
The blood pumping chamber is provided with a blood outflow duct 3 containing a suitable prosthetic outflow valve device 4. An inflow duct (not shown) with a suitable prosthetic inflow valve also is provided. An example of suitable mechanicalprosthetic valves are Bjork-Shiley valves although numerous other designs also may be employed. The blood pump inflow and outflow tracts preferably are connected respectively to known types of atrial cuffs 5 and arterial grafts (not shown) by snap-onquick-connect fittings 6 of any suitable design which facilitate surgical implantation of the TCP. The cuffs and grafts preferably are anastomosed to the atrial remanent and the aorta or pulmonary artery before the blood pumps are connected thereto.
The blood pumping membrane 2 preferably is of the single layer type formed from a high strength elastomeric biocompatible material. Polyurethane-based polymers such as Biomer and Avcothane are among the suitable materials for this application. These types of materials have been shown to inhibit high endurance and reliability in blood pumping operations. It is also important that the membrane of the blood pump exhibit low adhesion of thrombus and low generation of thrombo-emboli. The housingis formed of a suitable rigid metallic or plastic material, such as stainless steel coated with polyurethane or other biocompatible coatings, or glass or carbon fiber reinforced plastic. Typically, all internal surfaces of the blood pumps are coatedwith a suitable biocompatible material.
A suitable blood pump for use in the TCP of the present invention should be capable of providing a range of cardiac outputs of from 2.8 to about 9.5 liters per minute employing full stroke volume and at a beat rate of from about 35 to 120 beatsper minute.
In the TCP of the present invention, the above-described blood pumps are hydraulically actuated. While any incompressible fluid which is compatible with the actuator system components can be employed, the preferred actuation fluid is physiologicsaline solution (0.9 g percent NaCl) which is very close in saline composition to blood plasma. The use of saline as an actuation fluid promotes osmotic equilibrium and permits maintenance of a fixed actuation fluid inventory. It also eliminates theproblems associated with the use of certain other actuation fluids such as silicone oils, including diffusion of these oils into the body or diffusion and mixing of body fluids into the actuation fluid which can cause degradation of the polymer materialsin the flexible membrane.
The actuation system of the present invention in its most basic form comprises four basic components: (1) an actuation fluid reservoir or compliance sac, (2) actuation fluid pumping means, (3) a ventricular actuation chamber and (4) a three-wayventricular dump valve. The basic manner of operation of this system involves the pumping of actuation fluid from the actuation fluid reservoir into the ventricular actuation chamber to displace the flexible blood pump membrane and expel blood from thepump. The ventricular dump valve serves both to close off the actuation chamber outlet during expulsion of blood, and to drain or dump the actuation chamber fluid inventory at the end of expulsion, which permits refilling of the blood pumping chamber.
FIG. 3 schematically illustrates a preferred actuation system for the TCP of the invention. The blood pumps include a right ventricle RV and a left ventricle LV. Right ventricle RV is defined by a rigid housing 110 which is divided by aflexible bladder 112 into a blood pumping chamber 114 and a fluid actuation chamber 116. Similarly, left ventricle LV is defined by a rigid housing 210 which is divided by a flexible bladder 212 into a blood pumping chamber 214 and a fluid actuationchamber 216. Valve blood inlets 113, 213 and valved blood outlets 115, 215 interconnect the blood pumping chambers 114, 214 with the appropriate blood vessels.
Portions of the blood pump housings 110 and 210 are surrounded by a flexible membrane 10 which defines a fluid containing reservoir or compliance sac 12. This compliance sac faces the lung and other soft tissues in the thoracic cavity andcontains actuation fluid maintained at normal intrathoracic pressure levels. During operation, the fluid is dumped from each ventricle into the compliance sac during diastole and is removed from the compliance sac during systole. In the preferred formof operation, the ventricles are alternately actuated so as to minimize the change in volume of hydraulic actuation fluid in the system and therefore the overall size of the compliance sac. Copulsatile operation can, of course, be effected if desired.
In the preferred embodiment of FIG. 3, each of the blood pumps is independently actuated by its own three-way valve and pumping means. The three-way valve alternately directs actuation fluid from the pumping means into the actuation chamberduring systole, and then from the actuation chamber to the compliance sac during diastole. During diastole the valve blocks the flow of fluid from the pumping means. The valve can be power-operated by an electromechanical device such as a motor orsolenoid, or operated by forces which vary as a function of actuation fluid flow into the actuation chamber.
As used herein, the term "three-way" applies only to valves which in one state direct fluid from one of three distinct locations to a second of the three locations, and in a second state direct fluid from the second to the third of the threelocations. Such valves therefore channel fluid along only a single flow path which begins at the location of origin and terminates at the location of destination. No other flow path is provided in the valve which simultaneously carries fluid betweenany other locations. Of course, the single flow path referred to can comprise a plurality of channels which collectively carry the fuid from the location of origin to the location of destination. Examples of three-way valves are described below.
The pumping means preferably comprises a high speed, mixed flow, rotary pump driven by a brushless DC motor, although other suitable motor and pump designs may, of course, be used. The pump and motor bearings are totally immersed in andlubricated by the saline actuation fluid. In the preferred embodiment this hydraulic pump is designed to operate at a speed of about 7,000 to 15,000 rpm during the ventricular ejection phase (systole). During diastole, the pump can continue to run atthe same speed if the three-way valve is a power-operated type, or can be can be slowed to no less than about 1,000 to 1,200 rpm. At this low speed the pump will use less energy when pumping against the closed three-way powered valve, or will notproduce sufficient fluid flow to actuate a valve of the flow-responsive type, yet will maintain a full lubricating fluid film on the bearings.
The operation of the electric motor which drives the hydraulic pumps is continuously controlled in a manner hereinafter described, preferably using back emf commutation of the type described in Chambers et al, "Development of an Electrical EnergyConverter for Circulatory Devices," NTIS Publication No. PB-245 042, May, 1975, incorporated herein by reference.
Referring to FIG. 3, actuation pump 120, driven by motor 121, draws fluid from compliance sac 12 through a flexible duct 122 and delivers it to actuation chamber 116 via a flexible duct 124 through inlet 126. Similarly, actuating pump 220 drivenby motor 221 draws fluid from compliance sac 12 through a flexible duct 222 and delivers it to actuation chamber 216 via flexible duct 224 through inlet 226. Three-way valves 130, 230 are provided adjacent inlets 126, 226, respectively, for controllingthe inflow of actuation fluid into actuating chambers 116, 216 from pumps 120, 220, and the outflow of actuation fluid from actuating chambers 116, 216 into compliance sac 12 through outlets 132, 232. In the operational state shown in FIG. 3, pump 220is delivering actuation fluid to actuating chamber 216 through open inlet 226. Valve 230 has closed outlet 232, thereby preventing the escape of actuator fluid from actuating chamber 216. As pump 220 continues to operate, the volume of fluid withinactuating chamber 216 increases, thereby compressing blood pumping chamber 214 and forcing blood outwardly therefrom through outlet 215 into the systemic vasculature. At the same time, valve 130 blocks inlet 126 and leaves outlet 132 open, therebyallowing actuation fluid to drain from actuating chamber 116 into compliance sac 12. This drain of actuating fluid is caused by the right atrial blood pressure, which forces blood into the right ventricular blood pumping chamber 114 through inlet 113. Filling of each blood pumping chamber is therefore passive, as in the natural heart. When valve 130 moves to block oulet 132 and open inlet 126, and valve 230 moves to block inlet 226 and open outlet 232, the left ventricle is permitted to fill withblood, while the right ventricle is compressed to eject blood into the pulmonary system through outlet 115. Of course, the valves and pumps could be operated to actuate the blood pumps in copulsatile fashion, as long as a sufficient quantity ofactuation fluid is present.
FIG. 4 illustrates a preferred mechanical configuration for the actuation system embodiment shown schematically in FIG. 3. Three-way dump valves 330 are mounted on ventricle housings 110, 210. Flexible conduits 122, 124, 222, 224 connect thepumps 120, 220 to dump valves 330 and to the compliance sac 12.
Utilization of the above-described preferred embodiment provides a number of significant advantages. Utilization of two independent actuation systems allows for independent ventricular control as well as pump and motor optimization for eachventricle to maximize efficiency. Ducting losses are minimized by mounting the dump valves on the ventricle housings.
Other forms of TCP actuation schemes are schematically illustrated in FIGS. 5 through 8 and 16. In each of these figures, the right and left ventricular blood pumps are respectively designated RV, LV. The three-way dump valve associated witheach ventricle is designated V. The compliance sac or reservoir is designated R. Actuator pumps are designated P, while motors for driving the pumps are designated M.
FIG. 5 illustrates a configuration which is similar to that of FIGS. 3 and 4, with the exception that the separate actuator pumps are together driven by a single motor. The use of a single motor for driving two pumps effects a reduction inoverall TCP weight, although some flexibility of operation is sacrificed. This system can operate in alternately pulsatile fashion, with the motor running continuously and pulsation effected by actuation of power-operated three-way valves. Each pumpcan be selected to match the flow requirements for its respective ventricle, and motor speed can be varied to optimize fluid flow during systole in each ventricle. Copulsation also is possible, the three-way valves being either flow-responsive (with themotor stopped or slowed during diastole) or power-operated.
The embodiment illustrated in FIG. 6 is similar to that of FIG. 5, with the exception that a single pump and motor is used to supply actuation fluid to both ventricles. This configuration effects a further reduction of total TCP weight. Foralternate pulsation using power-operated three-way valves, motor speed can be varied to suit differing systolic requirements. Copulsation also is possible, the three-way valves being either flow-responsive (with the motor stopped or slowed duringdiastole) or power-operated.
In the embodiment of FIG. 7, the three-way dump valves are mounted together with their respective pumps and motors at locations remote from the ventricles. The ducts which connect the three-way valves to their respective ventricles musttherefore carry fluid in two directions: from the valve to the ventricle during systole, and from the ventricle to the valve during diastole. In this configuration, the blood pumping assembly, which consists only of the right and left ventricles, issmaller and lighter than those which have integrally mounted dump valves. Hence, the intrathoracic space required for the blood pumping assembly is minimized, although some efficiency may be sacrificed due to ducting losses. This system is functionallysimilar to that of FIGS. 3 and 4.
FIG. 8 illustrates an embodiment similar to that of FIG. 7, but wherein a single pump and motor are used to supply actuation fluid to both ventricles. This system is functionally similar to that of FIG. 6.
FIG. 16 illustrates a configuration which utilizes a reversible pump and check valves (a, b, c, d) in conjunction with the three-way ventricular dump valves to effect alternate pulsation only. During systole in the left ventricle and diastole inthe right, the pump draws fluid from the reservoir through check valve b and delivers it to the left ventricle through check valve d, while fluid drains from the right ventricle to the reservoir through the three-way dump valve. Check valve c preventsbackflow into the reservoir, while check valve a positively prevents backflow through the duct leading to the right ventricular three-way valve. A similar operation occurs during systole in the right ventricle and diastole in the left.
Check valves a and d are actually optional. In practice, the ventricular dump valve (flow responsive or power-operated) which is in its diastolic state (draining fluid from the ventricle to the reservoir) and substantially prevents back flowtowards the pump creates a sufficiently high impedance in the duct between the pump and the diastolic ventricle, as compared to the impedance in the valved duct between the reservoir and the pump, to resist any backflow to the pump.
The reversible pump is preferably of the axial flow type. Such a pump is highly efficient when operating under design conditions. For a given flow, the inertia of the pump impeller is minimized. In contrast with most positive displacementreversible pumps, such as gear pumps, there are no contacting parts other than the bearings. In sum, reversibility, low inertia, efficiency, and smooth, balanced operation are the salient features which makes an axial flow pump suitable for thisconfiguration.
The axial flow pump is preferably of the type which, like the mixed flow rotary pump used in the other configurations described above, is driven by a brushless DC motor, with the pump and motor bearings immersed in and lubricated by the salineactuation fluid. Back emf commutation is the preferred method of motor control. An example of a reversible axial flow pump suitable for use in the TCP according to the invention is illustrated and described in Jarvik, "The Total Artificial Heart,"Scientific American, Vol. 244, No. 1, pp. 74-80, January, 1981, which is incorporated by reference.
FIGS. 9, 10 and 11 illustrate one form of three-way dump valve which may be used in the cardiac prosthesis of the invention. Only one valve is illustrated, associated with a single ventricle. It is to be understood, however, that in a TCPhaving two ventricles, a three-way valve of this type would be associated with each ventricle.
The three-way valve 330 is mounted directly on a ventricular blood pump 310 of the type described above. Valve 330 is a power-driven, rotary plug valve having a plug 332 which is mounted for rotation within a generally cylindrical valve body334. Plug 332 has an arcuate groove 336 cut in its periphery so as to define, with the interior of valve body 334, a movable flow channel for directing fluid from one of three locations to another of the three locations. Rotary movement of plug 332 iseffected by an electric servomotor or electromagnetic actuator 338 which is attached to the end of valve body 334 and is operatively coupled to plug 332 in a suitable known manner. Motor or actuator 338 is designed to move plug 332 to one or the otherof two predetermined positions, described below, when it receives an appropriate electrical pulse from the control circuitry.
Valve body 334 includes an intake port or slit 340. An inlet manifold 342 having an inlet nipple 344 is attached to valve body 334 and communicates with inlet port 340. Nipple 344 is adapted to be connected to a duct which delivers actuationfluid to valve 330 from an actuator pump.
The interior of valve body 334 communicates with the actuation chamber within ventricle 310 through a port 346 and nipple 348. Nipple 348 also serves to secure valve 330 to ventricle 310 by engagement with ventricular nipple 350.
An outlet 352 is formed in the wall of valve body 334 through which actuation fluid is dumped from ventricle 310 into the compliance sac (not shown). During systole, with plug 332 in the position shown in solid lines in FIG. 11, actuation fluidpasses through nipple 344, manifold 342 and valve body 334 into the actuation chamber within ventricle 310. During diastole, with plug 332 in the position shown in broken lines in FIG. 11, actuation fluid is free to pass from the actuation chamberwithin ventricle 310 through valve body 334 and out through outlet 352 into the compliance sac.
Another form of power-driven three-way ventricular dump valve is illustrated in FIGS. 12 through 15. Valve 430 is mounted on ventricle 410 and is electromagnetically actuated by a solenoid 412. The actuation chamber within ventricle 410 isprovided with a large rectangular actuation fluid outlet 414 and three elongated rectangular actuation fluid inlets 416. The flow of actuation fluid through inlets 416 and outlet 414 is controlled by a finger valve assembly 432.
Finger valve assembly 432 has two sets of depending D-shaped tines or fingers 434, 436 which are respectively secured to arms 438, 440. Arms 438 and 440 are interconnected by a pivot arm 442 which is pivotally secured to ventricle 310 bybrackets 444 or the like. Pivot arm 442 is provided with a lateral extension 446 which is connected to the reciprocating actuating rod 413 of solenoid 412 by a pin 448. Solenoid 412 is secured to ventricle 410 beneath pivot arm 442 and between arms 438and 440.
A housing or cover 450 overlies arm 438 and fingers 434, and is in sealing engagement with ventricle 410 and pivot arm 442. Cover 450 has three outlet slits 452 which align with fingers 434. Cover 450 therefore forms an outlet manifold. Acover or housing 454 overlies arm 440 and fingers 436, and is sealed to ventricle 410 and around pivot arm 442. Cover 454 has a single inlet port 456 to which a duct 458 is connected for conveying actuation fluid from the actuator pump to ventricle 410. Cover 454 therefore forms an inlet manifold. The compliance sac (not shown) surrounds at least housing 450 so that actuation fluid emerging through slits 452 is dumped directly into the compliance sac.
During systole, solenoid rod 413 is moved to an upward position in response to an appropriate electrical signal from the controller. In this mode, finger valve assembly 432 is pivoted away from ventricle 410 such that, as shown in solid lines inFIG. 15, fingers 436 uncover inlets 416 and fingers 434 cover outlet slits 452. Actuation fluid enters housing 454 and flows into the actuation chamber within ventricle 410, while closed outlet slits 452 prevent the escape of actuation fluid. Duringdiastole, solenoid rod 413 is moved downwardly to pivot finger valve assembly 432 in the opposite direction so that fingers 434 and 436 occupy the positions shown in broken lines in FIG. 15. In this mode, inlets 416 are blocked, and outlet slits 452 areopened to permit actuation fluid to flow between fingers 434 and exit into the compliance sac.
While the above description refers to a single three-way dump valve associated with a single actuation chamber, it will, of course, be recognized that two or more dump valves of the same or different configurations can be associated with a singleactuation chamber. While the above description refers generally to independently operable dump valves associated with each actuation chamber of a TCP, it is, of course, possible to utilize dump valves which operate in a mutually dependent manner byproviding suitable coupling means.
Control and Operation
The above-described TCP system, especially in the preferred embodiments thereof, is particularly advantageous in its ability to respond to electronic controls which are designed to cause operation of the TCP in a manner consistent with naturalheart operation in humans. Basically, a modified Frank-Starling mechanism is the sole means of blood pump response to the physiological needs of the implant recipient. Each blood pump ejects whatever blood fills it, and as a result, the atrial pressureis related to cardiac output in the manner similar to the Frank-Starling response of the normal heart. Under the Frank-Starling mechanism, cardiac output is equal to venous return. Since cardiac output is equal to the heart rate times stroke volume,changes in cardiac output can be achieved either by changing the heart rate or the stroke volume. It is preferred according to the present invention to keep the stroke volume constant and achieve changes in cardiac output by changing heart rate. Constant stroke volume may be achieved by intrinsic means, e.g., by use of a stroke-limiting diaphragm, or by providing extrinsic control of diaphragm excursions in a known manner.
In the preferred embodiment of FIG. 3, heart rate control is effected by delivering controlled intermittent pulses of actuation fluid to the actuation chamber in response to a control signal supplied to each three-way ventricular dump valve andactuation fluid pump motor by control circuit C (FIG. 1). This control signal causes the ventricular dump valves to close and open appropriate ports in the manner described above at the beginning and end of systole, and controls pump motor speed. Thiscontrol signal can be generated in response to any of a number of known measurable variables which provide information that can be employed to effect blood pump operation to ensure that the physiologic requirements of blood flow are met. One suchvariable is atrial pressure, which can be measured in a known manner using pressure transducers. Landis et al, "Long-Term In Vivo Automatic Electronics Control of the Artificial Heart," Trans. ASAIO, Vol. XXIII, pp. 519-525, 1977, herein incorporatedby reference.
In the preferred manner of operating the TCP of the preseent invention, ventricular ejections will alternate thereby conserving actuation fluid. As indicated previously, however, copulsatile operation can be effected with a concomitant increasein the hydraulic reservoir capacity. One of the significant advantages of the preferred TCP design of the present invention is that by the use of separate actuation mechanisms for each ventricle, separate and optimal control of each ventricle can beachieved.
In the preferred embodiment power to energize the electric motors and control electronics is furnished by electromagnetic induction across the intact skin of the patient. Telemetry signal for indicating system and patient information are alsotransmitted in this manner, albeit usually in the opposite direction. This type of power supply and telemetering system has been described in the prior art and consists generally of a high frequency coupling transformer which includes a small flatinternal coil implanted subcutaneously and a larger flat external coil which is mounted over the implanted coil. The external coil can be carried in a vest, belt or other article of clothing. In practice, this energy transmission system can tolerateconsiderable movement of the external coil relative to the internal coil without adversely affecting the transfer of power or information into or out of the system. Power to drive the external coil can be provided in an external electronics pack whichcan contain batteries, battery charging electronics and other electronic systems useful in the monitoring of system and patient functions. Also included in the monitoring electronics are provisions for audible or visible alarms which warn of incipientmalfunction or problems.
The external battery pack can be designed to provide several hours of mobility for the patient. This battery pack when depleted can be easily replaced with a fully charged one. It can also be charged from AC line current when the patient is"plugged into the wall" or while the patient is using another battery pack, or from an automotive DC battery while motoring.
An implanted internal battery pack will also provide a temporary source with complete freedom from any external power. This will permit the recipient to undertake acts such as bathing and the like and will provide sufficient time to changeexternal power sources, i.e., changing the vest or external battery packs. In the preferred embodiment this internal battery will be located subcutaneously for easier periodic replacement.
While certain specific embodiments of the invention have been described with particularly herein, it should be recognized that various modifications thereof will appear to those skilled in the art. Therefore, the scope of the invention is to belimited solely by the scope of the claims appended hereto.